FutureHouse
FutureHouse是指一个非营利组织,它刚刚发布了四个超人类的AI科学家智能体,包括Crow(乌鸦)、Falcon(猎鹰)、Owl(猫头鹰)和Phoenix(凤凰)。这些智能体专门用于科学研究,已经通过了严格的基准测试,在搜索精度和准确性上超越了目前顶级的搜索模型,如o3-mini、GPT-4.5、Claude-3.7。此外,它们在直接文献搜索任务中,检索和综合能力比博士水平的研究人员更高。这
AI Engineer MCP 2025
MCP remote server for AI Engineer World's Fair 2025
LegoGPT
卡内基梅隆大学的研究团队开发出一款名为 LegoGPT 的 AI 模型,能够根据文字指令生成可实际搭建的乐高设计。 比如输入文本「基本款沙发」,一眨眼的功夫,乐高沙发就拼好了。 团队训练了一种自回归大型语言模型,通过预测下一个 token 的方式,判断下一块该放置什么积木。团队还为模型增加了有效性校验和带有物理感知的回滚机制,确保生成的设计不会出现积木重叠或悬空等问题,也就是说最终结果始终可行
Picture To Text
Convert images to text easily
InOtherWord - Translate Complex Docs
Expert translation service for complex documents and multi-format support.
AI Blood Test Analysis
AI-powered platform for analyzing blood test results.
File ChatAI
"AI Text Insights is a powerful platform for analyzing and summarizing various files in any language."
ChatGPT Desktop App - FaqAI
Effortlessly manage files, fetch website content, execute commands, and query databases using FileWork's intuitive platform.
WhatLetter
Translate, understand, and converse in your language.
Redactable
AI software for automated document redaction.
FetchFox
AI powered web scraper
Flowsecure.io
AI-powered survey tool for businesses.
Mindgrasp
Mindgrasp AI creates accurate notes and quizzes from different content types to aid learning and productivity.
DocuTranslate
Platform for official document translation
panda{·}etl (YC W24)
Automates document workflows for data extraction and analysis.
FormToExcel
Convert forms into Excel with high accuracy.
PDF AI Sheet
Google Sheets add-on for bulk PDF information extraction.
stirling PDF
Stirling-PDF是一款基于 Docker 的本地托管、功能强大的 Web PDF 处理工具。它支持您对 PDF 文件执行各种操作,包括拆分、合并、转换、重组、添加图像、旋转、压缩等等。这款本地托管的 Web 应用程序功能强大,功能全面,可满足您所有的 PDF 需求。 所有文件和 PDF 要么仅存在于客户端,要么仅在任务执行期间驻留在服务器内存中,要么仅在任务执行期间临时驻留在某个文件
Docses
Simplify complex legal jargon.
Scrap.so
AI-powered data collection assistant.
Website2GPT
Transforms websites into GPT-ready training data.
GenRPT
AI-driven platform for simplifying data analysis and visualization.
MSQA
MSQA(Multi-modal Situated Question Answering)是大规模多模态情境推理数据集,提升具身AI代理在3D场景中的理解与推理能力。数据集包含251K个问答对,覆盖9个问题类别,基于3D场景图和视觉-语言模型在真实世界3D场景中收集。MSQA用文本、图像和点云的交错多模态输入,减少单模态输入的歧义。引入MSNN(Multi-modal Next-step Navi
Apify
Comprehensive platform for web scraping and automation tools.
只显示前20页数据